|
Research question 5: How far can the advantages of "analog computing" beBog'liq Alireza FasihResearch question 5: How far can the advantages of "analog computing" be
used/gained through an emulation of analog computing on
digital hardware platforms like FPGA?
The main advantage of analog computing is that we can “nearly” simultaneously get the
result of complex mathematical differential equations. All signals are generated in parallel
and in real-time and the electronic components do compute simultaneously [30-32]. In the
traditional analog computing method one does face a scaling problem for the dynamic
range of the computing process. The dynamic range of all components is limited and one
additionally faces issues related to noise and high voltage. Therefore, is not possible to
compute any dynamic range and one therefore has to always rescale the ranges
appropriately. Another disadvantage of classical analog computing is that due to the fact
that the solutions appear in real-time one cannot easily record them for further analysis
purposes.
We can cover/overcome all of these problems by using the advantages of digital systems
like FPGA. Traditional analog computers are using operational amplifier based circuits to
model “addition”,” subtraction”, “multiplication” and “integration”
[32, 33]. These functions
are the basis for analog computing. Therefore, since we can realize all of these functions by
a digital circuit an emulation of “analog computing” on FPGA app
ears possible. We do use
the functional block “Digital Differential Analyzer (DDA)” to compute the integral of a
function over time. To get more speedup we can use fixed-point arithmetic calculation
instead of a floating-point one. Therefore, we must thoroughly check the ranges of values
and levels of accuracy for
assigning reserve bits for both “integer” and “fraction” parts. By
emulating analog computing on digital hardware platform like FPGA we become capable of
getting the solution in real-time without any limitation (contrarily to traditional analog
computing) related to scaling voltages. The reason is that instead of voltages we are
dealing with registers and fixed-point operators on data. A further advantage is that we do
not need any additional converter for/before storing data into the memory. We have full
control on the clock rate and this also represents a great advantage of using FPGA and a
digital architecture. Concerning setup time, debugging effort and configurability the digital
emulation of “analog computing” is clearly very superior to the traditional “real electronics
components” based analog computing systems.
|
| |
